CHIRLA is a multi-camera video dataset for person re-identification (Re-ID) and tracking research released by the Institute of Computing at the University of Alicante in 2025. The related paper results are "CHIRLA: Comprehensive High-resolution Identification and Re-identification for Large-scale Analysis", which aims to evaluate the performance of Re-ID and tracking algorithms in long-term and complex scenarios.

This dataset contains 596,345 video frames and 963,554 bounding boxes with identity annotations. The data was collected over a seven-month period using seven cameras, covering 22 identities and 70 video sequences. The images have a resolution of 1080×720 and a frame rate of 30 fps. The footage is captured in interconnected indoor environments, covering multiple time periods and environmental variations. All video frames are semi-automatically annotated, constructing a large-scale, multi-view, multi-time span multi-person tracking and re-identification dataset.